analyzing RNA seq

[kjaja]

Hi All,

I have RNA sequencing data (expression data) for a control sample and cases (with disease) and at multiple time points but no replicates of. any ideas on how to analyze this type of data?

thanks

[Richard Finney]

check out Deseq2 and Dexseq.

[crazyhottommy]

No replicates is a bad experimental design.
Next maSigPro: updating maSigPro Bioconductor
package for RNA-seq time series

and here http://www.hindawi.com/journals/bmri/2013/203681/

[mbblack]

No replicates means you have no ability to use parametric statistics to assess significance (other than using permutations or some other iterative approach to simulate=guestimate variance).

You should have a look at some alternative analysis. Two that come to mind is a simple Rank Product analysis of fold change lists (which you can do in R or in a simple Excel spreadsheet for that matter). There is also GFOLD (http://bioinformatics.oxfordjournals...8/21/2782.long ) which uses a baseyian posterior probablity estimate of the distribution of fold changes to derive statistics and ranking. GFOLD will be helped by replicates (as they allow for a more robust estimate of the posterior probability distribution) but is valid without them (the only assumption then is that read count for any gene is poisson distributed).

GFOLD is available in BioConductor.